The Mayo Clinic Alzheimer's Disease Patient Registry (ADPR) was established in 1986 and is involved in community-based studies of aging and dementia. The retrospective studies using the Mayo Clinic medical records-linkage system have provided incidence and prevalence rates and survival data for dementia and Alzheimer's disease (AD) over a 30-year period. The accumulated data base is one of the few resources worldwide where secular trends in prevalence, incidence and survival can be studied. The ADPR has also evaluated several risk and protective factors. The prospective studies involving longitudinal cohorts of subjects with AD, other dementias, and mild cognitive impairment, as well as a matched set of control subjects, have been for 11 years. In the present application we propose to continue to investigate incidence, prevalence, and survival for dementia and AD and to expand this work into vascular dementia. Secular trends in the oldest-old segment of the population will be studied. We also propose several case-control studies of putative protective factors such as estrogen, non-steroidal anti-inflammatory, and H2 blocker drugs. The prospective studies will address the issue of the boundary between normal aging and mild cognitive impairment. Using results derived from the retrospective studies, the prospective projects will be designed to determine early predictors of a subsequent cognitive decline cognitive decline. Three approaches will be used to address this theme. 1) Pairs of cases of cognitively impaired individuals (AD, other dementias, mild cognitive impairment) and age- and gender-matched control subjects will be recruited at a rate of 50-60 pairs per year. This will provide for continuity with previous and ongoing studies. 2) The normal control cohort will be expanded to a total of 500 subjects over the five years of the proposed grant period to allow sufficient power to evaluate multiple potential predictor variables for the development of cognitive impairment. 3) We will expand our recruitment of subjects in the oldest-old segment of the population (age 90-99 years) to approximately 500 subjects since this group offers a unique opportunity to study aging and cognitive impairment with a high likelihood of obtaining postmortem examinations. In all three approaches outlined above, we will obtain clinical, neuropsychological, biological, and neuroimaging variables and correlate them with postmortem findings. The Mayo ADPR provides the opportunity to study aging and dementia in a well-defined community through the use of the medical records-linkage system and a well-established longitudinal prospective cohort. The ability to translate research findings from retrospective medical record studies to the prospective cohort is unique. We hope to address important issues concerning aging and cognitive impairment through the proposed studies.